RESUMEN
Intensity-modulated radiation therapy (IMRT) has been widely used in treating head and neck tumors. However, due to the complex anatomical structures in the head and neck region, it is challenging for the plan optimizer to rapidly generate clinically acceptable IMRT treatment plans. A novel deep learning multi-scale Transformer (MST) model was developed in the current study aiming to accelerate the IMRT planning for head and neck tumors while generating more precise prediction of the voxel-level dose distribution. The proposed end-to-end MST model employs the shunted Transformer to capture multi-scale features and learn a global dependency, and utilizes 3D deformable convolution bottleneck blocks to extract shape-aware feature and compensate the loss of spatial information in the patch merging layers. Moreover, data augmentation and self-knowledge distillation are used to further improve the prediction performance of the model. The MST model was trained and evaluated on the OpenKBP Challenge dataset. Its prediction accuracy was compared with three previous dose prediction models: C3D, TrDosePred, and TSNet. The predicted dose distributions of our proposed MST model in the tumor region are closest to the original clinical dose distribution. The MST model achieves the dose score of 2.23 Gy and the DVH score of 1.34 Gy on the test dataset, outperforming the other three models by 8%-17%. For clinical-related DVH dosimetric metrics, the prediction accuracy in terms of mean absolute error (MAE) is 2.04% for D 99 , 1.54% for D 95 , 1.87% for D 1 , 1.87% for D mean , 1.89% for D 0.1 c c , respectively, superior to the other three models. The quantitative results demonstrated that the proposed MST model achieved more accurate voxel-level dose prediction than the previous models for head and neck tumors. The MST model has a great potential to be applied to other disease sites to further improve the quality and efficiency of radiotherapy planning.
RESUMEN
BACKGROUND: Most partial nephrectomies (PNs) are performed with hilar occlusion to reduce blood loss and optimize visualization. However, the histologic status of the preserved renal parenchyma years after PN is unknown. OBJECTIVE: To compare the histologic chronic kidney disease (CKD) score of renal parenchyma before and years after PN, and to explore factors associated with CKD-score increase and glomerular filtration rate (GFR) decline. DESIGN, SETTING, AND PARTICIPANTS: A retrospective review of 147 renal cell carcinoma patients who underwent PN and subsequent radical nephrectomy (RN) due to tumor recurrence was performed in 19 Chinese centers and Cleveland Clinic. Macroscopic normal renal parenchyma was evaluated at least 5 mm away from the tumor in PN specimens and at remote sites in RN specimens. INTERVENTION: PN/RN and ischemia. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Histologic CKD score (0-12) represents a summary of glomerular/tubular/interstitial/vascular status. Predictive factors for a substantial increase of CKD score (≥3) were evaluated by logistic regression. RESULTS AND LIMITATIONS: Sixty-five patients with all necessary data were analyzed. The median interval between PN and RN was 2.4 yr. Median durations of warm ischemia (n = 42) and hypothermia (n = 23) were both 23 min. The histologic CKD score was increased after RN in 47 (72%) patients, with 29 (45%) experiencing more substantial increase (≥3). There was no significant difference in the change of CKD score related to the type and duration of ischemia (p = 0.7 and p = 0.4, respectively) or interval from PN to RN (p > 0.9). However, patients with comorbidities of hypertension, diabetes, and/or pre-existing CKD (hypertension [HTN]/diabetes mellitus [DM]/CKD) demonstrated increased rate and extent of CKD-score increase. On univariate analysis, HTN/DM/CKD was the only predictor of a substantial CKD-score increase (odds ratio: 3.53 [1.12-11.1]). Decline of GFR was modest and similar between patients with/without a substantial CKD-score increase. CONCLUSIONS: Within the context of conventional, limited durations of ischemia, histologic deterioration of preserved parenchyma after PN appears to be primarily due to pre-existing medical comorbidities rather than ischemia. A subsequent decline in renal function was mild and independent of histologic changes. PATIENT SUMMARY: After clamped PN, the preserved renal parenchyma demonstrated histologic deterioration in many cases, which correlated with the presence of comorbidities such as hypertension, diabetes mellitus, or chronic kidney disease. In contrast, the type and duration of ischemia did not correlate with histologic changes after PN, suggesting that ischemia insult had only limited impact on parenchyma deterioration.